The relationship between characterological depression, or DSM-III-R primary early-onset dysthymia (PEOD), and the major affective and personality disorders has long been debated. The PI's preliminary work suggests that PEOD is a complex condition with links to both the major mood and personality disorders. The present proposal seeks to extend this work with a clinical and family study of PEOD. Subjects will include 100 outpatients with PEOD, 50 outpatients with episodic major depression (EMD), and 50 normal controls, and their first-degree relatives. Four major questions will be addressed: a. Is PEOD a more severe form of mood disorder than EMD from the standpoint of familial transmission? b. Which personality disorders are more common in PEOD than EMD? c. Other than mood disorders, are there other familial factors which distinguish PEOD from EMD? d. Which of five models of the relationships between PEOD and the major affective and Axis II disorders best explains the pattern of mood and personality disorders in relatives of PEOD, EMD, and normal probands?